All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jesper Dangaard Brouer <brouer@redhat.com>
To: Ryan Goodfellow <rgoodfel@isi.edu>
Cc: "xdp-newbies@vger.kernel.org" <xdp-newbies@vger.kernel.org>,
	brouer@redhat.com
Subject: Re: zero-copy between interfaces
Date: Sat, 18 Jan 2020 11:14:05 +0100	[thread overview]
Message-ID: <20200118111405.28fd1c75@carbon> (raw)
In-Reply-To: <20200117175409.GC69024@smtp.ads.isi.edu>

On Fri, 17 Jan 2020 12:54:09 -0500
Ryan Goodfellow <rgoodfel@isi.edu> wrote:

> On Mon, Jan 13, 2020 at 06:04:11PM +0100, Jesper Dangaard Brouer wrote:
> > On Mon, 13 Jan 2020 10:28:00 -0500
> > Ryan Goodfellow <rgoodfel@isi.edu> wrote:
> >   
> > > On Mon, Jan 13, 2020 at 12:41:34PM +0100, Jesper Dangaard Brouer wrote:  
> > > > On Mon, 13 Jan 2020 00:18:36 +0000
> > > > Ryan Goodfellow <rgoodfel@isi.edu> wrote:
> > > >     
> > > > > The numbers that I have been able to achive with this code are the following. MTU
> > > > > is 1500 in all cases.
> > > > > 
> > > > >     mlx5: pps ~ 2.4 Mpps, 29 Gbps (driver mode, zero-copy)
> > > > >     i40e: pps ~ 700 Kpps, 8 Gbps (skb mode, copy)
> > > > >     virtio: pps ~ 200 Kpps, 2.4 Gbps (skb mode, copy, all qemu/kvm VMs)
> > > > > 
> > > > > Are these numbers in the ballpark of what's expected?    
> > > > 
> > > > I would say they are too slow / low.
> > > > 
> > > > Have you remembered to do bulking?
> > > >     
> > > 
> > > I am using a batch size of 256.  
> > 
> > Hmm...
> > 
> > Maybe you can test with xdp_redirect_map program in samples/bpf/ and
> > compare the performance on this hardware?  
> 
> Hi Jesper,
> 
> I tried to use this program, however it does not seem to work for bidirectional
> traffic across the two interfaces?

It does work bidirectional if you start more of these xdp_redirect_map
programs.  Do notice this is an example program.  Look at xdp_fwd_*.c
if you want a program that is functional and uses the existing IP route
table for XDP acceleration.

My point is that there are alternatives for doing zero-copy between
interfaces... A xdp_redirect_map inside the kernel out another
interface is already zero-copy.

I'm wondering why did you choose/need AF_XDP technology for doing forwarding?

-- 
Best regards,
  Jesper Dangaard Brouer
  MSc.CS, Principal Kernel Engineer at Red Hat
  LinkedIn: http://www.linkedin.com/in/brouer

  reply	other threads:[~2020-01-18 10:14 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-01-13  0:18 zero-copy between interfaces Ryan Goodfellow
2020-01-13  9:16 ` Magnus Karlsson
2020-01-13 10:43   ` Toke Høiland-Jørgensen
2020-01-13 15:25     ` Ryan Goodfellow
2020-01-13 17:09       ` Toke Høiland-Jørgensen
2020-01-14  7:47         ` Magnus Karlsson
2020-01-14  8:11           ` Toke Høiland-Jørgensen
2020-01-13 15:11   ` Ryan Goodfellow
2020-01-14  9:59     ` Magnus Karlsson
2020-01-14 20:52       ` Ryan Goodfellow
2020-01-15  1:41         ` Ryan Goodfellow
2020-01-15  7:40           ` Magnus Karlsson
2020-01-15  8:20             ` Magnus Karlsson
2020-01-16  2:04               ` Ryan Goodfellow
2020-01-16 14:32                 ` Magnus Karlsson
2020-01-17  9:45                   ` Magnus Karlsson
2020-01-17 17:05                     ` Ryan Goodfellow
2020-01-21  7:34                 ` Magnus Karlsson
2020-01-21 13:40                   ` Maxim Mikityanskiy
2020-01-22 21:43                     ` Ryan Goodfellow
2020-01-27 14:01                       ` Maxim Mikityanskiy
2020-01-27 15:54                         ` Magnus Karlsson
2020-01-30  9:37                           ` Maxim Mikityanskiy
2020-01-30  9:59                             ` Magnus Karlsson
2020-01-30 11:40                               ` Magnus Karlsson
2020-02-04 16:10                                 ` Magnus Karlsson
2020-02-05 13:31                                   ` Magnus Karlsson
2020-02-06 14:56                                     ` Maxim Mikityanskiy
2020-02-07  9:01                                       ` Magnus Karlsson
2020-01-17 17:40         ` William Tu
2020-01-13 11:41 ` Jesper Dangaard Brouer
2020-01-13 15:28   ` Ryan Goodfellow
2020-01-13 17:04     ` Jesper Dangaard Brouer
2020-01-17 17:54       ` Ryan Goodfellow
2020-01-18 10:14         ` Jesper Dangaard Brouer [this message]
2020-01-18 14:08           ` Ryan Goodfellow
2020-01-26  4:53             ` Dan Siemon
2020-01-17 12:32 ` Björn Töpel
2020-01-17 12:32   ` [Intel-wired-lan] " =?unknown-8bit?q?Bj=C3=B6rn_T=C3=B6pel?=
2020-01-17 17:16   ` Ryan Goodfellow
2020-01-17 17:16     ` [Intel-wired-lan] " Ryan Goodfellow
2020-01-17 18:10     ` Ryan Goodfellow
2020-01-17 18:10       ` [Intel-wired-lan] " Ryan Goodfellow
2020-01-20  8:24     ` Magnus Karlsson
2020-01-20  8:24       ` [Intel-wired-lan] " Magnus Karlsson
2020-01-20 18:33       ` Ryan Goodfellow
2020-01-20 18:33         ` [Intel-wired-lan] " Ryan Goodfellow
2020-01-20 17:04     ` Björn Töpel
2020-01-20 17:04       ` [Intel-wired-lan] " =?unknown-8bit?q?Bj=C3=B6rn_T=C3=B6pel?=

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200118111405.28fd1c75@carbon \
    --to=brouer@redhat.com \
    --cc=rgoodfel@isi.edu \
    --cc=xdp-newbies@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.